AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Errors-in-variables Fr\'echet Regression with Low-rank Covariate Approximation

Dogyoon Song, Kyunghee Han

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Statistical analysis of ML practice · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply singular value thresholding (SVT, low-rank approximation) instead of naive EIVglobal Frechet regression · Gaussian errors, p=150, n=100MSPE: 1.267 -> 0.686Improved
apply SVT instead of naive EIVglobal Frechet regression · Gaussian errors, p=600, n=400MSPE: 1.242 -> 0.657Improved
apply SVT instead of naive EIVglobal Frechet regression · Laplacian errors, p=300, n=200MSPE: 1.521 -> 0.677Improved
apply SVT instead of naive EIV, smallest gapglobal Frechet regression · Gaussian errors, p=600, n=100MSPE: 1.062 -> 1.054Improved
use SVT on error-prone covariates instead of oracle REF on error-free covariatesglobal Frechet regression · Gaussian errors, p=150, n=100MSPE: 1.613 -> 0.686Improved
use naive EIV on error-prone covariates instead of oracle REFglobal Frechet regression · Gaussian errors, p=150, n=100MSPE: 1.613 -> 1.267Improved
use oracle REF instead of SVTglobal Frechet regression · Gaussian errors, p=150, n=100bias: 0.312 -> 0.107Improved
use SVT instead of REF on varianceglobal Frechet regression · Gaussian errors, p=150, n=100sqrt(Var): 1.147 -> 0.531Improved
use oracle REF instead of SVT on in-sample fitglobal Frechet regression · Gaussian errors, p=300, n=100MSE: 0.217 -> 0.000Improved
increase covariate dimension p from 150 to 600SVT · Gaussian errors, n=100MSPE: 0.686 -> 1.054Hurt
increase sample size n from 100 to 400SVT · Gaussian errors, p=150MSPE: 0.686 -> 0.412Improved
change error distribution from Gaussian to Laplaciannaive EIV · p=150, n=100MSPE: 1.267 -> 1.538Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy